The US AI Action Plan prioritises global dominance over safety, risking market concentration and weaker regulations worldwide. Open models and interoperable standards can spread AI benefits more equitably while safeguarding public interest